Document not found! Please try again

Reversible Addition− Fragmentation Chain Transfer (RAFT

Jul 2, 2009 - polymerization kinetics and the living character of RAFT-IMEP, ... preparing doubly hydrophilic block copolymers under appropriate RAFT/...
0 downloads 0 Views 2MB Size
Macromolecules 2009, 42, 5043–5052

5043

DOI: 10.1021/ma900715c

Reversible Addition-Fragmentation Chain Transfer (RAFT) Polymerization in an Inverse Microemulsion System: Homopolymerization, Chain Extension, and Block Copolymerization† Atsushi Sogabe‡ and Charles L. McCormick*,‡,§ ‡ Department of Polymer Science and §Department of Chemistry and Biochemistry, The University of Southern Mississippi, Hattiesburg, Mississippi 39406

Received April 2, 2009; Revised Manuscript Received June 9, 2009

ABSTRACT: We report the first successful reversible addition-fragmentation chain transfer inverse microemulsion polymerization (RAFT-IMEP). The inverse microemulsion (IME) system conditions were optimized by generating pseudo-three-component phase diagrams. The IME consisted of the hydrophilic monomer N,N-dimethylacrylamide (DMA), water, hexanes, nonionic surfactants, and a cosurfactant. The polymerization kinetics and the living character of RAFT-IMEP, conducted with varying amounts of dispersed aqueous phase, were similar to that observed for aqueous RAFT polymerization of DMA. The colloidal stability during polymerization was also investigated via dynamic light scattering. Decreasing the weight fraction of the dispersed aqueous phase leads to an increase in colloidal stability and a decrease in microemulsion size. An increase in the dispersity of molecular weight is also observed with a decrease in the weight fraction of dispersed aqueous phase. This observation is attributed to uncontrolled polymer that forms during the early stages of RAFT-IMEP. Chain extension or block copolymer formation is accomplished by simple addition of DMA or DEA to the polyDMA macroCTA, demonstrating the potential utility for preparing doubly hydrophilic block copolymers under appropriate RAFT/IMEP conditions.

† Paper no. 140 in a series on Water-Soluble Polymers. *To whom correspondence should be addressed. E-mail: Charles. [email protected].

eventually phase separate. Microemulsions, on the other hand, are generated spontaneously without the necessity of vigorous agitation and are thermodynamically stable. However, it is often difficult to find conditions for generating stable microemulsions, since the type and relative concentrations of surfactants, cosurfactant, continuous phase, and other components must be carefully determined. Since the discovery of RAFT, many researchers have attempted polymerization in emulsion systems.7,18-29 The first reported attempt of RAFT macroemulsion polymerization was by Le et al., obtaining polystyrene with molecular weight of 53 200 and a PDI of 1.37.18 Later work utilizing RAFT macroemulsion polymerization revealed problems of rate retardation, phase separation, and loss of livingness.20,21,25,28,30 These difficulties have been attributed to the diffusion of the chain transfer agent (CTA) or oligomeric CTA into and out of the emulsion particles.25,31 Furthermore, the ratio of monomer to CTA concentration in the macroemulsion is not constant during polymerization due to differences in their respective diffusion coefficients.32 To limit the diffusion of the CTA, seeded emulsion,33,34 starved fed,35 oligomeric CTA,26,36 and ab initio emulsion polymerization techniques27,37,38 have been implemented. However, these techniques can be laborious since each usually involves multistep reactions. Because of the difficulties associated with RAFT macroemulsion polymerization, several groups have turned to miniemulsion systems since each monomer droplet can, in principle, serve as a locus of polymerization without the diffusion of monomer and CTA.19,22,39,40 In some instances, phase separation of oligomeric CTA occurs, which has been termed “superswelling”.20,39 The reported problems associated with RAFT macro- and miniemulsion polymerizations have led to investigations of the alternative microemulsion systems which should have thermodynamic stability.41

r 2009 American Chemical Society

Published on Web 07/02/2009

Introduction The recent progress in controlled/“living” radical polymerization techniques has provided impetus for synthesizing welldefined polymers with advanced architectures. Such techniques include nitroxide-mediated polymerization (NMP),1 atom transfer radical polymerization (ATRP),2 and reversible additionfragmentation chain transfer (RAFT) polymerization.3-6 Among these techniques, RAFT polymerization is arguably the most versatile since it is compatible with a wide variety of functional monomers under mild reaction conditions.7,8 The robust nature of the RAFT process has allowed significant progress in the area of aqueous RAFT polymerization.5,6 Our group, having a long-standing interest in water-soluble functional polymers, has recently focused on controlled polymerization of anionic,9 cationic,10,11 zwitterionic,12 and neutral13,14 monomers utilizing aqueous RAFT polymerization. Emulsion polymerization is often used in large-scale production of polymers due to the facile control of reaction kinetics, ease of product isolation, and subsequent application. Emulsion systems are generally classified as one of the following types: macroemulsion (the most common), miniemulsion, and microemulsion. These systems differ fundamentally in the dimensions of the dispersed phase. While the diameters of the dispersed phase in macroemulsions are typically greater than 1 μm,15 miniemulsions (50-500 nm)16,17 and microemulsions (10-100 nm)15 are on the nanometer scale. Although mini- and microemulsions overlap in the size of the dispersed phase, these two systems are significantly different in terms of stability. Miniemulsions are generated by vigorous stirring, followed by ultrasonication and are thermodynamically unstable microphase systems that will

pubs.acs.org/Macromolecules

5044

Macromolecules, Vol. 42, No. 14, 2009

Sogabe and McCormick Scheme 1

The RAFT polymerizations discussed above are all based on oil-in-water emulsions and are not applicable for synthesizing hydrophilic polymers. Compared to the oil-in-water emulsion systems, a minimal number of inverse emulsion systems have been reported.42-50 Our literature survey reveals only one report regarding RAFT inverse miniemulsion polymerization51 and none regarding RAFT inverse microemulsion polymerization (RAFT-IMEP). For the former, experimental molecular weights matched with the theoretical values at conversions less than 40% using a hydrophilic initiator. At higher conversion, experimental molecular weights were greater than the predicted values, and this is attributed to the hydrolysis of CTA. Moreover, the colloidal stability of the system was not discussed. In this report, we describe, to our knowledge, the first successful RAFT-IMEP. The inverse microemulsion system was obtained by carefully constructing pseudo-three-component phase diagrams. This system includes DMA as the hydrophilic monomer, nonionic surfactants, and a cosurfactant. The kinetic behavior and the living character of RAFT-IMEP, with systematic variation of the dispersed aqueous phase, were compared to those of RAFT aqueous solution polymerization. The colloidal stability of the microemulsion during polymerization was also investigated via dynamic light scattering. Chain extension with DMA and block copolymerization with DEA were successfully performed from the polyDMA macroCTA, indicating the utility of this method in preparing block copolymers including doubly hydrophilic and/or amphiphilic systems. Experimental Section Materials. N,N-Dimethylacrylamide (DMA), N,N-diethylacrylamide (DEA), and N,N-dimethylpropionamide (DMPA) (Aldrich, Milwaukee, WI) were distilled under reduced pressure. Polyoxyethylene(3) oleyl ether (POE(3)C18), polyoxyethylene (6) oleyl ether (POE(6)C18) (Nihon Emulsion, Tokyo, Japan), and hexanes (Aldrich, Milwaukee, WI) were used as received. 2,20 -Azobis[2-(2-imidozolin-2-yl)propane] dihydrochloride (VA-044) was used as the initiator (gifts from Wako Chemicals USA, Inc.) The difunctional CTA 2-(1-carboxy-1-methylethylsulfanylthiocarbonylsulfanyl)-2-methylpropionic acid (CMP) was used without further purification (gift from Noveon, Inc.).

Preparation of Pseudo-Three-Component Phase Diagrams. The following three components were used to generate phase diagrams. The oil component consisted of hexanes, the surfactant component consisted of 33.3 wt % of POE(3)C18 and 66.7 wt % of POE(6)C18, and the aqueous component consisted of (a) DI water, (b) 20 wt % DMA, (c) 20 wt % DMPA, or (d) 20 wt % DMA and 20 wt % DMPA and a small amount of CMP. These three components were mixed at various ratios and left at 40 °C for 2 h. The phase diagrams were generated from visual inspection and conductivity measurements of the resulting mixtures. RAFT Polymerization of DMA in Inverse Microemulsion. General Procedure (Scheme 1). Aqueous Component. CMP (14.2 mg, 5.05 mmol) was dissolved in DMA (2.00 g, 20.2 mmol) and DMPA (2.00 g, 19.8 mmol) in a 20 mL vial. DI water (5.00 g) was then added to the solution and cooled to 0 °C. A stock solution cooled to 0 °C containing VA-044 (3.26 mg, 1.01 mmol) and DI water (1.00 g) was added to the vial. The solution was stirred and purged with N2 for 1 h at 0 °C. Mixture of Oil Component and Surfactant Component. Hexanes (8.75 g), POE(3)C18 (0.700 g), and POE(6)C18 (1.40 g) were added to a 20 mL round bottle flask. The solution was stirred and purged with N2 for 1 h at 25 °C and then cooled to 0 °C. Preparation of Inverse Microemulsion (IME) and RAFT Polymerization. The aqueous component (1-4 g) was added to the oil and surfactant mixture at 0 °C under N2. The solution was purged with N2 for another 1 h and then was immersed oil bath at 40 °C. The inverse microemulsion was generated within 10 min after heating. The reaction was terminated after 20 h by cooling to 0 °C in an ice bath followed by exposure to air. The polymerization kinetics were monitored by taking aliquots at predetermined time intervals utilizing degassed syringes. Monomer concentration was determined by UV absorbance at 260 nm. GPC was used to determine the molecular weight at different time intervals while UV absorbance at 310 nm was utilized to detect the CTA moiety. Chain Extension in a One-Pot System. The polyDMA with target DP = 100 was synthesized (10 h, ∼100% conversion) by RAFT-IMEP with an 8.5 wt % aqueous component as described above. An additional monomer solution containing DMA (0.20 g, 2.02 mmol) and DMPA (0.20 g, 1.98 mmol) was prepared and purged with N2 at 0 °C for 1 h. This second

Article

Macromolecules, Vol. 42, No. 14, 2009

batch of monomer was added into the IME mixture under N2. The chain extension was conducted for another 10 h at 40 °C to yield a final target DP of 200. Block Copolymerization in a One-Pot System. The polyDMA with target DP = 100 was synthesized (10 h, ∼100% conversion) by RAFT-IMEP with an 8.5 wt % aqueous component as described above. DEA (0.272 g, 2.14 mmol) as the second monomer was purged with N2 at 0 °C for 1 h. This second batch of monomer was added into the IME mixture under N2. The block copolymerization was conducted for another 6 h at 40 °C to yield a final target of poly(DMA50-DEA106-DMA50). Gel Permeation Chromatography. After the polymerization, hexanes were removed from the mixture by purging with N2 gas. Ethanol was added to dissolve the mixture, and then surfactants were separated by using centrifugal filtration (MWCO 3000, Ultracel YM-3 or MWCO 10 000 Ultracel YM-10). Following filtration, the samples were analyzed by aqueous size exclusion chromatography (ASEC) using an eluent of 20%/80% acetonitrile/0.05 M Na2SO4 at a flow rate of 0.3 mL/min at 25 °C. This instrument was equipped with TOSOH Biosciences TSK-GEL columns [Supre AW3000 G3000 PWXL (